RK3399PRO 运行OpenCL 算子出错

使用3399ProMaliT860GPU执行一些并行运算，现遇到一个问题,求助一下坛友：
在Debian10环境下，使用OpenCL实现并行运算算子，发现在执行 clEnqueueNDRangeKernel之后，程序会“Aborted”
使用得失最简单的矩阵相加测试程序（见附件程序），我们在其他平台上（PC及其他ARM板子上运行是正常）。

程序运行时候的dump信息如下：

Starting program: /share/TestCL/main.app 
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/aarch64-linux-gnu/libthread_db.so.1".
OpenCL 1.2 v1.r18p0-01rel0.5630b190419266e7fe8b09ec0007fb39
[New Thread 0x7fb5e2b1b0 (LWP 4427)]
[New Thread 0x7fb562a1b0 (LWP 4428)]
[New Thread 0x7fb4e291b0 (LWP 4429)]
[New Thread 0x7fb46281b0 (LWP 4430)]
[New Thread 0x7fb3e271b0 (LWP 4431)]
[New Thread 0x7fb36261b0 (LWP 4432)]
[New Thread 0x7fb2e251b0 (LWP 4433)]
[New Thread 0x7fb26241b0 (LWP 4434)]
[New Thread 0x7fb1e231b0 (LWP 4435)]
Line: 223
Thread 9 "mali-cmar-backe" received signal SIGABRT, Aborted.
[Switching to Thread 0x7fb26241b0 (LWP 4434)]
__GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
50    ../sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt full
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
        set = {__val = {0, 15, 15, 0, 0, 5, 0, 1, 548547829760, 548453627008, 
            548513712272, 548515074048, 548453630384, 548512566312, 56, 0}}
        pid = <optimized out>
        tid = <optimized out>
        ret = <optimized out>
#1  0x0000007fb5f708e8 in __GI_abort () at abort.c:79
        save_stage = 1
        act = {__sigaction_handler = {sa_handler = 0x3, sa_sigaction = 0x3}, 
          sa_mask = {__val = {3, 0, 0, 6066178765026259759, 
              7020319398736196453, 123627969080937, 0, 0, 0, 0, 0, 
              4616194021471028225, 4616194021471028225, 0, 0, 25769803782}}, 
          sa_flags = 6, sa_restorer = 0x7fb7f51b00 <abort@got.plt>}
        sigs = {__val = {32, 0 <repeats 15 times>}}
#2  0x0000007fb665012c in ?? () from /lib/aarch64-linux-gnu/libOpenCL.so  #最终的错误应该是libOpenCL.so内部
No symbol table info available.
#3  0x0000007fffffb200 in ?? ()
No symbol table info available.
Backtrace stopped: previous frame identical to this frame (corrupt stack?)

其他一些使用到的库版本信息：
OpenCL： V1.2
Mali GPU 驱动： r18p0
系统： Debian 10
其他库版本信息： linux-vdso.so.1
libOpenCL.so=>/lib/aarch64-linux-gnu/libOpenCL.so
libstdc++.so.6=>/lib/aarch64-linux-gnu/libstdc++.so.6
libm.so.6 => /lib/aarch64-linux-gnu/libm.so.6
libgcc_s.so.1=>/lib/aarch64-linux-gnu/libgcc_s.so.1
libc.so.6 => /lib/aarch64-linux-gnu/libc.so.6
libpthread.so.0=>/lib/aarch64-linux-gnu/libpthread.so.0
libdl.so.2=>/lib/aarch64-linux-gnu/libdl.so.2
librt.so.1=>/lib/aarch64-linux-gnu/librt.so.1
libdrm.so.2=>/lib/aarch64-linux-gnu/libdrm.so.2
libtoybrick_gbm.so.0=> /lib/aarch64-linux-gnu/libtoybrick_gbm.so.0
libwayland-client.so.0=> /lib/aarch64-linux-gnu/libwayland-client.so.0
libwayland-server.so.0=> /lib/aarch64-linux-gnu/libwayland-server.so.0
libffi.so.6=>/lib/aarch64-linux-gnu/libffi.so.6